[time-nuts] Line Voltage [Was: Anyone (ideally in the UK) ...]

Dr. David Kirkby (Kirkby Microwave Ltd) drkirkby at kirkbymicrowave.co.uk
Sun Jan 1 05:36:50 EST 2017

On 1 Jan 2017 05:07, "Chris Albertson" <albertson.chris at gmail.com> wrote:
> Just a comment for anyone who wants to log line voltage v. time.   If
> you have an APC "Smart UPS" battery backup unit these will log voltage
> and frequency to a file.

If the information is for your own use only, that is probably fine.  But if
you intend challenging the electricity supply company over it, I would use
a laboratory multimeter with a valid calibration certificate. It will give
your measurements far more credibility than the data from a UPS.

I would suggest that a 5.5 digit laboratory multimeter with a current
certificate from Keysight would be better than an 8.5 digit 3458A that is
out of the calibration period by a day, despite the latter having lower

The point where you measure the voltage is probably important too.
Obviously wire resistance in a ring main reduces the voltage, so there's
not much point reporting that the voltage is low unless measured at the
point of the incoming supply. I suspect that it is better to measure there
whatever ones cause  for complete is. Possibly a motor run internally could
act as a generator and push the supply above the incoming voltage.

I don't know what (if any) averaging should be done.  Does one sample every
cycle?  I would probably set my 3457A  to sample 10 power line cycles (200
ms) here in the UK.  I think collecting data every cycle would be a bit
excessive, but maybe not.  One can always post process the data later to do
some averaging.

What's the sample rate on your APC UPS?


More information about the time-nuts mailing list